IWO (improved walrus optimizer) was proposed to address the shortages of slow convergence rate and easy to fall into local optima in WO (walrus optimizer). Firstly, an adaptive factor was introduced to the male walrus using Halton sequence for position update to increase the optimization accuracy of the algorithm. Secondly, a reverse learning strategy based on multiple centers of gravity was introduced into the position update formula of female walruses to make full use of the diversity of the population and find the global optimum. Numerical experiments were conducted on 23 benchmark test functions to compare with other advanced algorithms to verify the superiority of IWO. The algorithm is also applied to solve the coverage optimization of WSN (wireless sensor network) and optimization of reinforced concrete beam and column structure. Experimental results show that, compared with other advanced optimization algorithms, the algorithm IWO has higher search accuracy and convergence speed on the test functions. The coverage performance of the IWO algorithm has been significantly improved In WSN coverage optimization, and the coverage rate can reach 98.87%. IWO algorithm reduces the total cost of reinforced beam column structures by 14.31%, compared to WO, which verifies its better optimizing performance and practicality.